search

McGranahanLab / CONIPHER-wrapper

8 stars 2 forks source link

Issue about Error in simpleClusterList[[as.character(ITH1clust)]] #4

Open neuroniko opened 11 months ago

neuroniko commented 11 months ago

Dear TracerX team : I'd like to thank you for creating CONIPHER, it has been immensely useful in my work. I've been working with constructing tumour phylogenetic trees and came across an issue that I'm hoping you might be able to help me with. In most case, CONIPHER will work with this code: Rscript $Rscript_path --case_id ${case_id} --working_dir ${clustering_dir} --script_dir ${scriptDir} --input_tsv ${inputTSV} --clean_clusters TRUE --min_cluster_size 3 --nProc 1 However, in somes cases, after pyclone step, I encountered such error: Iteration: 9900 Number of clusters: 5 DP concentration: 707.759699054 Beta-Binomial precision: 912.87700996

PyClone build_table --config_file /public/workspace/guosy/project/WES/Wu_bam_file/LC_First/WuYY$ 105131_ZSX_TP/CONIPHER/1.Clustering//Wu_YY_2105131_ZSX_TP_cluster9//Wu_YY_2105131_ZSX_TP_cluster9$ config.yaml --table_type old_style --out_file /public/workspace/guosy/project/WES/Wu_bamfile/LC$ irst/Wu_YY_2105131_ZSX_TP/CONIPHER/1.Clustering//Wu_YY_2105131_ZSX_TP_cluster9//Wu_YY_2105131_ZSX$ TP.results.tsv --max_clusters 1 --burnin 1000 Error in simpleClusterList[[as.character(ITH1clust)]] : attempt to select less than one element in get1index Execution halted

Any guidance or suggestions would be greatly appreciated.

Thank you for your time and your contribution to the CONIPHER.

arihueb commented 9 months ago

Hi, it is hard to say based on the error alone, but these a few possible issues it could be related to:

  1. No truncal cluster: if your samples do not have shared mutations that are present in all samples this will create issues.
  2. Have you performed 'force calling' of mutations, i.e. assigned read counts for each mutation in each tumor sample even if the mutation is absent? (See Supplementary Methods of https://doi.org/10.1038/s41596-023-00913-9 as well as Methods of https://doi.org/10.1038/s41586-023-05783-5).
  3. Are there low frequency mutations that are not defined as present in any region? The new version of CONIPHER should filter out such a cluster.

We have recently updated the tool and wrapper (CONIPHER v2.2.0), so I would recommend installing the latest conda environment and trying to rerun the analysis.